Tissue microarrays (TMAs) are a cost-effective tool to study biomarkers in clinical research. Cervical cancer (CC) is one of the most prevalent in women worldwide, with the highest prevalence in low-middle-income countries due to a lack of organized screening. CC is associated with persistent high-risk human papillomavirus infection. Several biomarkers have been studied for diagnostic, therapeutic, and prognostic purposes. We aimed to evaluate and validate the effectiveness of TMA in CC compared to whole slide images (WSs). We selected and anonymized twenty cases of CC. P16, cytokeratin 5 (CK5), cytokeratin 7 (CK7), programmed death-ligand 1 (PD-L1), and CD8 expression were immunohistochemically investigated. All WS were scanned and 10 representative virtual TMA cores with 0.6 mm diameter per sample were selected. Ten random combinations of 1-5 cylinders per case were assessed for each biomarker. The agreement of scoring between TMA and WS was evaluated by kappa statistics. We found that three cores of 0.6 mm on TMA can accurately represent WS in our setting. The Kappa value between TMA and WS varied from 1 for p16 to 0.61 for PD-L1. Our study presents an approach to address TMA sampling that could be generalized to TMA-based research, regardless of the tissue and biomarkers of interest.